SWE-Bench AI Task Auditor - Freelance AI Trainer Project
Core
Review software engineering tasks used to train and assess advanced AI models to ensure technical accuracy, realism, reproducibility, and appropriate testing.
Role type
Freelance AI trainer / AI task auditor
Builds
AI training workflows and evaluation datasets
Domain
Artificial Intelligence / Software Engineering
Deliverable
production ML models
Required skills
software engineering judgment, codebase integration analysis, test failure investigation, logic issue identification, technical accuracy verification, reproducibility assessment
Preferred skills
experience evaluating AI systems, practical software engineering expertise
Technologies
AI training frameworks, codebases, testing suites
Responsibilities
Review software engineering tasks for technical accuracy and realism, investigate codebase integrations and test failures, identify logic issues and technical challenges, apply practical software engineering judgment to evaluate AI training data
Seniority
Experienced, hands-on IC